Analysis and Projection of Temperature and Rainfall 20110217130224 0

Scientific Basis:
Analysis and Projection of Temperature and Rainfall

AUTHORS
Indonesia Climate Change Sectoral Roadmap – ICCSR
Scientific Basis: Analysis and Projection of Temperature and Rainfall
Adviser
Prof. Armida S. Alisjahbana, Minister of National Development Planning/Head of Bappenas
Editor in Chief
U. Hayati Triastuti, Deputy Minister for Natural Resources and Environment, Bappenas
ICCSR Coordinator
Edi Effendi Tedjakusuma, Director of Environmental Affairs, Bappenas
Editors
Irving Mintzer, Syamsidar Thamrin, Heiner von Luepke, Dieter Brulez
Synthesis Report
Coordinating Authors for Adaptation: Djoko Santoso Abi Suroso
Scientific Basis: Analysis and Projection of Temperature and Rainfall Report
Author: Tri Wahyu Hadi
Technical Supporting Team
Chandra Panjiwibowo, Hendra Julianto, Leyla Stender, Tom Harrison, Ursula Flossmann-Krauss
Administrative Team
Altamy Chrysan Arasty, Risnawati, Rinanda Ratna Putri, Siwi Handinah, Wahyu Hidayat, Eko Supriyatno,
Rama Ruchyama, Arlette Naomi, Maika Nurhayati, Rachman
i
ICCSR - Scientific Basis: Analysis and Projection of Temperature and Rainfall
ACKNOWLEDGMENTS
The Indonesia Climate Change Sectoral Roadmap (ICCSR) is meant to provide inputs for the next five
year Medium-term Development Plan (RPJM) 2010-2014, and also for the subsequent RPJMN until
2030, laying particular emphasis on the challenges emerging in the forestry, energy, industry, agriculture,
transportation, coastal area, water, waste and health sectors. It is Bappenas policy to address these
challenges and opportunities through effective development planning and coordination of the work
of all line ministries, departments and agencies of the Government of Indonesia (GoI). It is a dynamic
document and it will be improved based on the needs and challenges to cope with climate change in the
future. Changes and adjustments to this document would be carried out through participative consultation
among stakeholders.
High appreciation goes to Mrs. Armida S. Alisyahbana as Minister of National Development Planning
/Head of the National Development Planning Agency (Bappenas) for the support and encouragement.
Besides, Mr. Paskah Suzetta as the Previous Minister of National Development Planning/ Head of
Bappenas who initiated and supported the development of the ICCSR, and Deputy Minister for Natural
Resources and Environment, Ministry of National Development Planning /Bappenas, who initiates and
coordinates the development of the ICCSR.
To the following steering committee, working groups, and stakeholders, who provide valuable comments
and inputs in the development of the ICCSR Scientific Basis for Analysis and Projection of Temperature
and Rainfall document, their contributions are highly appreciated and acknowledged:
Steering Committee (SC)
Deputy of International Cooperation, Coordinating Ministry for Economy; Secretary of Minister,

Coordinating Ministry for Public Welfare; Executive Secretary, Agency for Meteorology, Climatology;
Deputy of Economy, Deputy of Infrastructures, Deputy of Development Funding, Deputy of Human
Resources and Culture, Deputy of Regional Development and Local Autonomy, National Development
Planning Agency; and Chief of Secretariat of the National Council for Climate Change.
Working Group
National Development Planning Agency
Sriyanti, Yahya R. Hidayat, Bambang Prihartono, Mesdin Kornelis Simarmata, Arum Atmawikarta,
Montty Girianna, Wahyuningsih Darajati, Basah Hernowo, M. Donny Azdan, Budi Hidayat, Anwar Sunari,
Hanan Nugroho, Jadhie Ardajat, Hadiat, Arif Haryana, Tommy Hermawan, Suwarno, Erik Amundito,
iii
Rizal Primana, Nur H. Rahayu, Pungki Widiaryanto, Maraita, Wijaya Wardhana, Rachmat Mulyanda,
Andiyanto Haryoko, Petrus Sumarsono, Maliki
Agency for Meteorology, Climatology and Geophysics
Edvin Aldrian, Dodo Gunawan, Nurhayati, Soetamto, Yunus S, Sunaryo
National Institute of Aeuronatics and Space
Agus Hidayat, Halimurrahman, Bambang Siswanto, Erna Sri A, Husni Nasution
Research and Implementatiton of Technology Board
Eddy Supriyono, Fadli Syamsuddin, Alvini, Edie P
National Coordinating Agency for Survey and Mapping
Suwahyono, Habib Subagio, Agus Santoso
Grateful thanks to all staff of the Deputy Minister for Natural Resources and Environment, Ministry of
National Development Planning/ Bappenas, who were always ready to assist the technical facilitation as
well as in administrative matters for the finalization process of this document.
The development of the ICCSR document was supported by the Deutsche Gesellschaft für Technische
Zusammenarbeit (GTZ) through its Study and Expert Fund for Advisory Services in Climate Protection
and its support is gratefully acknowledged.
iv
Remarks from Minister of National
Development Planning/Head of Bappenas
We have seen that with its far reaching impact on the world’s ecosystems
as well as human security and development, climate change has emerged
as one of the most intensely critical issues that deserve the attention
of the world’s policy makers. The main theme is to avoid an increase
in global average temperature that exceeds 2˚C, i.e. to reduce annual
worldwide emissions more than half from the present level in 2050.
We believe that this effort of course requires concerted international
response – collective actions to address potential conflicting national
and international policy initiatives. As the world economy is now facing
a recovery and developing countries are struggling to fulfill basic needs
for their population, climate change exposes the world population to
exacerbated life. It is necessary, therefore, to incorporate measures to
address climate change as a core concern and mainstream in sustainable development policy agenda.
We are aware that climate change has been researched and discussed the world over. Solutions have
been proffered, programs funded and partnerships embraced. Despite this, carbon emissions continue
to increase in both developed and developing countries. Due to its geographical location, Indonesia’s
vulnerability to climate change cannot be underplayed. We stand to experience significant losses. We will
face – indeed we are seeing the impact of some these issues right now- prolonged droughts, flooding and
increased frequency of extreme weather events. Our rich biodiversity is at risk as well.
Those who would seek to silence debate on this issue or delay in engagement to solve it are now
marginalized to the edges of what science would tell us. Decades of research, analysis and emerging
environmental evidence tell us that far from being merely just an environmental issue, climate change will
touch every aspect of our life as a nation and as individuals.
Regrettably, we cannot prevent or escape some negative impacts of climate change. We and in particular
the developed world, have been warming the world for too long. We have to prepare therefore to adapt
to the changes we will face and also ready, with our full energy, to mitigate against further change. We
have ratified the Kyoto Protocol early and guided and contributed to world debate, through hosting
the 13th Convention of the Parties to the United Nations Framework Convention on Climate Change
(UNFCCC), which generated the Bali Action Plan in 2007. Most recently, we have turned our attention
to our biggest challenge yet, that of delivering on our President’s promise to reduce carbon emissions by
26% by 2020. Real action is urgent. But before action, we need to come up with careful analysis, strategic
v
ICCSR - SCIentIfIC BaSIS: analySIS and PRojeCtIon of temPeRatuRe and RaInfall
planning and priority setting.
I am delighted therefore to deliver Indonesia Climate Change Sectoral Roadmap, or I call it ICCSR, with the
aim at mainstreaming climate change into our national medium-term development plan.
The ICCSR outlines our strategic vision that places particular emphasis on the challenges emerging in
the forestry, energy, industry, transport, agriculture, coastal areas, water, waste and health sectors. The
content of the roadmap has been formulated through a rigorius analysis. We have undertaken vulnerability
assessments, prioritized actions including capacity-building and response strategies, completed by
associated financial assessments and sought to develop a coherent plan that could be supported by line
Ministries and relevant strategic partners and donors.
I launched ICCSR to you and I invite for your commitment support and partnership in joining us in
realising priorities for climate-resilient sustainable development while protecting our population from
further vulnerability.
Minister for National Development Planning/

Head of National Development Planning Agency
Prof. Armida S. Alisjahbana
vi
Remarks from Deputy Minister for Natural
Resources and Environment, Bappenas
To be a part of the solution to global climate change, the government

of Indonesia has endorsed a commitment to reduce the country’s
GHG emission by 26%, within ten years and with national resources,
benchmarked to the emission level from a business as usual and, up to
41% emission reductions can be achieved with international support
to our mitigation efforts. The top two sectors that contribute to the
country’s emissions are forestry and energy sector, mainly emissions
from deforestation and by power plants, which is in part due to the fuel
used, i.e., oil and coal, and part of our high energy intensity.
With a unique set of geographical location, among countries on the

Earth we are at most vulnerable to the negative impacts of climate
change. Measures are needed to protect our people from the adverse effect of sea level rise, flood,
greater variability of rainfall, and other predicted impacts. Unless adaptive measures are taken, prediction
tells us that a large fraction of Indonesia could experience freshwater scarcity, declining crop yields, and
vanishing habitats for coastal communities and ecosystem.
National actions are needed both to mitigate the global climate change and to identify climate change
adaptation measures. This is the ultimate objective of the Indonesia Climate Change Sectoral Roadmap, ICCSR.
A set of highest priorities of the actions are to be integrated into our system of national development
planning. We have therefore been working to build national concensus and understanding of climate
change response options. The Indonesia Climate Change Sectoral Roadmap (ICCSR) represents our long-term
commitment to emission reduction and adaptation measures and it shows our ongoing, inovative climate
mitigation and adaptation programs for the decades to come.
Deputy Minister for Natural Resources and Environment

National Development Planning Agency
U. Hayati Triastuti
vii
TABLE OF CONTENTS
ACKNOWLEDGMENTS iii
Remarks from Minister of National Development Planning/Head of Bappenas v
Remarks from Deputy Minister for Natural Resources and Environment, Bappenas vii
TABLE OF CONTENTS viii
LIST OF TABLES xi
LIST OF FIGURES xii
LIST OF ABBREVIATION xiv
1. INTRODUCTION 1
1.1 Background and Objectives 2
1.2 Scope of Discussions and Contents of This Report 3
2. GENERAL DESCRIPTION 5
2.1 Global Climate System and Definition of Climate Change 6
2.2 Global Climate Change Scenario and Projection 7
2.3 Global Circulation Model (GCM) Performance in Tropical Area 9
2.4 Study of Climate Change and Its Impact in Indonesia 9
3. ANALYSIS AND PROJECTION OF CLIMATE CHANGE IN INDONESIA 13
3.1 Data Availability 14
3.1.1 Observational Data 14
3.1.2 GCM Output Data 14
3.1.3 High-Resolution GCM Output Data 15
3.2 Data Processing Method 16
3.2.1 Observational Data Processing 16
3.2.2 GCM Data Processing 19
viii
4. RESULTS AND DISCUSSIONS 21
4.1. Current Climate Change Trends in Indonesia 22
4.1.1 Rainfall Change 22
4.1.2 Surface Temperature Change 24
4.1.3 Probability Change for Extreme Weather and Climate 25
4.2 Climate Change in 2030s Period and 2080 Period 28
4.2.1 Projection of Surface Temperature Change 28
4.2.2 Projection of Rainfall Change 31
4.2.3 Projection of Probability Change for Extreme Weather and Climate 31
5. GENERAL RECOMMENDATION FOR SECTORAL ADAPTATION 33
5.1 Description of Climate Change for Each Region 34
5.1.1 Climate Change in Java-Bali Area 34
5.1.2 Climate Change in Sumatera Area 34
5.1.3 Climate Change in Kalimantan Area 35
5.1.4 Climate Change in Sulawesi Area 35
5.1.5 Climate Change in Nusatenggara Area 35
5.1.6 Climate Change in Moluccas Area 36
5.1.7 Climate Change in Papua Area 36
5.2 Additional Notes 36
5.2.1 Long Term Climate Variability 36
5.2.2 Climate Projection Uncertainties 38
6. CLOSING REMARK 41
References 43
APPENDIX A : RAINFALL BASELINE BASED ON GPCC DATA 45
APPENDIX B1 : RAINFALL CHANGE PROJECTION 2010-2015 47
APPENDIX F : STATION RAINFALL GRAPH 47
ix
APPENDIX B2 : RAINFALL CHANGE PROJECTION 2015-2020 51
APPENDIX C : AREA DIVISION 54
APPENDIX D1 : TEMPERATURE PROJECTION FOR 2030s AND 2080s IN JAVA BALI 56
APPENDIX D2 : TEMPERATURE PROJECTION FOR 2030s AND 2080s IN KALIMANTAN 57
APPENDIX D3 : TEMPERATURE PROJECTION FOR 2030s AND 2080s IN MALUKU 58
APPENDIX D4 : TEMPERATURE PROJECTION FOR 2030s AND 2080s IN NUSA TENGGARA 59
APPENDIX D5 : TEMPERATURE PROJECTION FOR 2030s AND 2080s IN PAPUA 60
APPENDIX D6 : TEMPERATURE PROJECTION FOR 2030s AND 2080s IN SULAWESI 61
APPENDIX D7 : TEMPERATURE PROJECTION FOR 2030s AND 2080s IN SUMATERA 62
APPENDIX E1 : RAINFALL PROJECTION FOR 2030s AND 2080s IN JAVA-BALI 63
APPENDIX E2 : RAINFALL PROJECTION FOR 2030s AND 2080s IN KALIMANTAN 64
APPENDIX E3 : RAINFALL PROJECTION FOR 2030s AND 2080s IN MALUKU 66
APPENDIX E4 : RAINFALL PROJECTION FOR 2030s AND 2080s IN NUSA TENGGARA 67
APPENDIX E1 : RAINFALL PROJECTION FOR 2030s AND 2080s IN PAPUA 68
APPENDIX E1 : RAINFALL PROJECTION FOR 2030s AND 2080s IN SULAWESI 69
APPENDIX E1 : RAINFALL PROJECTION FOR 2030s AND 2080s IN SUMATERA 70
APPENDIX F : STATION RAINFALL GRAPH 70
x
LIST OF TABLES
Table 2.1 SRES scenarios and emission level discussed in this study (adapted from IPCC, 2007b) 7
Table 3.1 hort description of GCM output data analysed in this study. 15
xi
LIST OF ABBREVIATION
AMO Atlantic Multidecadal Oscillation
ASEAN Association of South East Asian Nations
Bappenas National Development Planning Agency
BMKG Meteorology Climatology and Geophysics Agency
CDF Cumulative Distribution Function
CGCM Coupled General Circulation Model
CO2 Carbon Dioxide
CH4 Methane
EN El Nino
ENSO El Niño Southern Oscillation
FGD Focus Group Discussion
GCM General Circulation Model
GHCN Global Historical Climatological Network
GHG Greenhouse Gas
GPCC Global Precipitation Climatology Center
IOD Indian Ocean Dipole
IPCC AR-4 Intergovernmental Panel on Climate Change Assessment Report 4
LAPAN National Aeronautics and Space Agency
LN La Nina
MJO Madden-Julian Oscillation
MRI Meteorological Research Institute
NOAA National Oceanic and Atmospheric Agency
PDO Pacific Decadal Oscillation
PPM Part per Million
SRES Special Report on Emission Scenario
UNEP United Nations Environmental Program
UNFCCC United Nations Framework Convention on Climate Change
WMO World Meteorological Organization
xiv
1
INTRODUCTION
1
1.1 Background and Objectives
Since the IPCC (Intergovernmental Panel on Climate Change) was established in 1988 by WMO (World
Meteorological Organization) and UNEP (United Nations Environmental Program), and thereafter
published the reports of their studies, the global climate change issue has influenced the opinion of most of
the people in the world and started to affect development policies in many countries, including Indonesia.
The main problem of the global climate change issue is the increase of Earth surface temperature due
to the greenhouse effect caused by accumulated greenhouse gases such as CO2 (carbon dioxide) and
CH4 (methane) in the atmosphere. It is known that the concentration of atmospheric greenhouse gases
increases as a result of - among others - industrial activities, use of fossil fuels and land use, land use
change and forestry processes. Further increase of Earth surface temperature in the future, is predicted
to induce thermal expansion of ocean water and the melting of icecaps in both poles, North Pole and
South Pole (Greenland and Antarctic), that lead to the rise of global mean sea level. Furthermore, it is
assumed with increasing certainty that global warming will also influence or has already influenced the
global climate pattern that triggers the increasing frequency of extreme weather and climatic events.
Considering the abundance of negative impacts that are predicted to occur in the future, the international
community has made efforts to mitigate (by means of reduction of anthropogenic greenhouse gas
emissions) and to adapt to (by means of development strategies and policies to increase resilience or
enhance protection against the negative impacts) the global climate change. However, any mitigation or
adaptation action plan will create complex socio-economic problems as its implementations will directly
influence a broad spectrum of societies.
In general, adaptation and mitigation actions and plans devised in various international forums have been
based on IPCC climate projections which are formulated on the basis of studies using global climate
models. However, climate model output has its inherent uncertainties (IPCC, 2007) and, according to
Schneider (2002), the uncertainty propagates and grows from various levels: from the response of the
global climate system according to emission scenarios, to regional climate change scenario, and to the
possible impacts that may occur. The problem of uncertainties in climate models is still debated among
climate scientists but the decision to mitigate and adapt must still be taken regardless the discrepancy
between academic understanding and policy-making (Pielke, 2003).
Due to the uncertainties in long-term climate projection, climate change adaptation policies are facing
three possible risks (Australian Government, 2005) i.e. (1) over adaptation (excessive adaptation), (2)
under-adaptation (lacking of adaptation), or (3) mal-adaptation (wrong adaptation or adaptation that is
unnecessary). For this reason, to reduce the risks due to uncertainties, climate change adaptation action
plans need to be supported by a comprehensive understanding on the climatic change behaviour at
regional and local scales. Thus, a study of the Scientific Basis including an analysis of future climate
projections is needed for the formulation of a precise adaptation strategy to cope with climate change.
This report on the Analysis and Projection of Climate Change in Indonesia is part of the effort to devise
2
a Roadmap for Mainstreaming of Climate Change into National Development Planning (hereafter the
Roadmap for brevity). The objectives of this study are:
1) Identification of climate change in Indonesia, to gain scientific evidence on its characteristics and
various probable impacts on the national development.
2) Understanding of the general characteristics (pattern and magnitude) of future climate change
hazard in Indonesia so its impacts to the development sectors can be estimated, and
3) Preparing climate data that can be used to analyse the risk of climate change to the development
sectors with a temporal and spatial resolution that is tailored as best as possible with the unique
characteristics of each region (and/or sub-region).
With all shortcomings and numerous obstacles faced during its implementation, it is hoped that this study
can contribute significantly to the process of developing the Roadmap.
1.2 Scope of Discussions and Contents of This Report
This Analysis and Projection of Climate Change in Indonesia incorporates results from existing studies,
be it from individual studies or studies conducted by institutions such as BMKG (Agency for Meteorology,
Climatology, and Geophysiscs/Badan Meteorologi, Klimatologi, dan Geofisika), LAPAN (National Aeronautics
and Space Agency/ Lembaga Penerbangan dan Antariksa Nasional), etc. Therefore, this study was initialised
by conducting a Scientific Basis FGD (Focus Group Discussion) on February 4, 2009 meant to map
the products of climate change studies relevant to Indonesia that meet the previously mentioned three
objectives. However, we could only gather very limited information from the FGD. It was then necessary
for us to conduct data analysis in this study and to include data processing methods in the scope of
discussion. The contents of this report are outlined below.
The general description of review of the problems related to the analysis and projection of climate
change is discussed in Chapter 2. Whereas, Chapter 3 contains a description of methodology, data, and
data processing techniques used in the analysis and projection of climate change in Indonesia. The result
of the data analysis is discussed in Chapter 4, while Chapter 5 contains additional discussions related to
confidence level of the results and other notes on aspects of long-term climatology, as well as general
recommendation for sectors. In the last chapter we provide some suggestions on how to strengthen
academic study on climate change in Indonesia.
3
4
2
GENERAL DESCRIPTION
5
2.1 Global Climate System and Deﬁnition of Climate Change
Academically speaking in the broadest sense, climate can be defined as the state of the climate system
that comprises the components such as: atmosphere (air), hydrosphere (water), lithosphere (the solid part
of the Earth, including Earth crust, and volcanoes), cryosphere (ice sheets), and biosphere (vegetation).
All components of the climate system interact with one another to preserve the system and create a
particular condition in a particular time and space. The interactions in the climate system involve complex
processes with timescales ranging from seconds to millions of years and spatial dimensions that range
from molecular to planetary scales.
Climate can be defined as the statistical description of the elements of climate/weather such as
temperature, precipitation (rainfall), wind, etc., that at least describe the average value (mean) and its
variance for several decades (30 years by WMO’s definition). According to Meehl (2000), climate change
can be identified by comparing the probability distributions as illustrated in Figure 1 so that the changes
(between two 30-year consecutive periods) in the mean or variance, or in both values, could be identified.
Based on this definition, all changes that occur within each of the 30-year period are referred to as the
climate variability. Another consequence is that to identify the current climate change, we need historical
climate data for at least 60 years. However, this is not the only definition that is used internationally; in
the UNFCCC (United Nation Framework Convention on Climate Change), for example, climate change
is defined as the changes in climate that occur in addition to its natural variability over a similar timescale,
and it can be attributed either directly or indirectly to human activities that cause changes in global
atmosphere composition and/or land uses.
Figure 2.1 Identification of climate changes statistically from (a) change in mean, (b) change in
variance, and (c) changes in mean and variance. Vertical axis is probability and horizontal axis is
(climate) parameter value. (Adopted from Meehl, 2000).
6
In reality, when global climate change occurs, it is very difficult to scientifically distinguish between
natural changes and the changes attributed to human activities. Moreover, in the context of adaptation,
it is unnecessary and irrelevant to associate the impact of climate change with the cause of the climate
changes. For this reason, we adopt Meehl’s (2000) structure of thinking to formulate the identification
of climate change operationally (Figure 2.1) so that, in principle, climate change hazard can be identified
statistically without addressing the problem of whether a human (anthropogenic) factor is involved or
not.
2.2 Global Climate Change Scenario and Projection
IPCC has comprehensively discussed global climate projection problem based on GCM (Global
Circulation Model ) simulations as described in Climate Change 2007: The Physical Science Basis
(IPCC AR-4, The Fourth Assessment Report). In general, IPCC models mainly produce features of
climate change that are caused by increased concentrations of greenhouse gases in the atmosphere
resulting from anthropogenic actions and based on known mechanisms. In regard to that, the scenarios
of increasing greenhouse gases concentration become an inseparable part of climate change projection.
These scenarios are determined typically by some alternative assumptions of global economic and
population growth, technology development, etc. A widely used set of scenarios are described in the
SRES (Special Report on Emissions Scenarios) published by IPCC in 2000. Aside from SRES, there are
other scenarios like IS92 and SA90 that were used in previous IPCC reports. A detailed explanation of
these scenarios can be found for example in IPCC-TGICA (2007). To simplify the problem, in this study
we will only use IPCC data that are related to three SRES scenarios B1, A1B, and A2 that are selected
based on CO2 concentration level as shown in Table 2.1.
Table 2.1 SRES scenarios and emission level discussed in this study (adapted from IPCC, 2007b)
SRES Scenario Emission Category Stabilization in 2100

A2 High emission not stabilized
A1B Moderate emission 750 ppm
B1 Lowest emission in SRES 550 ppm
In relation to the three SRES scenarios B1, A1B, and A2, IPCC projected the range of increase of global
surface temperature as shown in Figure 2.2. IPCC AR-4 also contains a comprehensive analysis on global
and regional climate changes and the model data that can be accessed via internet (see more detailed
explanation below). In general, it is found that all models show an increase in surface temperature, in
spite of different values. As it is shown in Figure 2.2, in 2100 the global surface temperature will rise
1-4º C relative to the average temperature in 1980-1999. However, there are much less similarities in the
7
projected rainfall, as different models produce different rainfall values that vary significantly in space and
time. Aside from that, for the needs of a climate change impact study, a commonly known problem in
using GCM data is inadequate resolution to represent climate variability at any specific location. For this
reason, to fit the needs for impact study, a process called downscaling is needed.
In principle, downscaling is a technique to process the global model output to obtain a higher resolution
data. According to the tools being used, the methods for downscaling can generally be classified either
as (1) statistical or (2) dynamical downscaling. Dynamical downscaling is based on the application
of a mathematical model that is consistent with the physical description of the climate system. The
disadvantages of this method are that it requires a huge amount of computational resources and the
estimated values usually have large bias against observational data. On the other hand, the statistical
downscaling method can produce an estimation with sufficient accuracy using much less computational
resources but the results may not be able to be interpreted physically as the model parameters for different
variables may largely vary (models are less consistent). Furthermore, the implementation of statistical
downscaling requires a relatively complete observational data. In principle, both models can be used for
downscaling climate projection as both produce results with comparable accuracy (Wilby et al., 2004).
As previously discussed, the source of climate projection is the output of global model (GCM). Before
implementing any downscaling procedure, it is necessary to investigate whether the GCM output data
under consideration are good indicators for climatic variations in the studied region.
Figure 2.2 Global surface temperature change projection using three SRES scenario, B1, A1B and A2.
(Source: IPCC, 2007)
8
2.3 Global Circulation Model (GCM) Performance in Tropical Area
The climate of the Indonesian region is mainly controlled by the monsoon circulation system with two
basic patterns of rainy and dry seasons. Based on the pattern of the peak rainfall periods during rainy
season, there are three types of climate region (e.g. Aldrian and Susanto, 2003); they are: (1) monsoonal
(with one peak rainfall period around December-January-February), (2) equatorial (with two peaks around
April-May and October-November), and (3) local (one dominant peak around June-July). In addition to
monsoon, Indonesia is also climatologically influenced by the global ocean-atmosphere dynamics that
create various climate variability phenomena so that the rain characteristic for a given month can vary
greatly from year to year. The most significant interannual (with the period of occurrence between 2
and 5 years ) climate variability phenomena in Indonesia are associated with ENSO (El Nino Southern
Oscillation) in the Pacific Ocean and, more recently found, IOD (Indian Ocean Dipole) in the Indian
Ocean. Furthermore, there are also MJO (Madden-Julian Oscillation) phenomenon that can induce
rainfall variation in 30-60 days period and other phenomena that are still being studied by scientists.
The existence of various climatological phenomena in the tropical region leads to a highly complex
interaction with the monsoon circulation system and cause a variability that is difficult to be perfectly
simulated in climate models.
Several scientists have conducted studies to evaluate the performance of global climate model for South
East Asia region that is influenced by monsoon circulation system. It is found that various models produce
different results. Nevertheless, one of the important aspects studied in global model performance study
is the ENSO-Monsoon correlation and the correspondence between the model results and observation.
Joseph and Nigam (2006) reported that HadCM3, the model run in England, produces the most realistic
ENSO signal. Meanwhile, Annamalai et al. (2006) found that from 16 IPCC AR-4 models evaluated;
only 4 of them show a robust ENSO-Monsoon tele-connection. The four models are GFDL 2.0, GFDL
2.1 (USA), MRI (Japan), and ECHAM-5 (German). A (literature) study on IPCC model performance
was conducted by Smith (2009) from various existing validation results (not limited to tropical areas).
The result was a ranking of 22 models that is consistent with the previous studies. The top ten models
are: (1)HadCM3, (2)MIROC3.2(hires), (3)GFDL2.1, (4)GFDL2.0, (5)MIROC3.2(medres), (6)ECHO-
G, (7)HadGEM1, (8)ECHAM5, (9)MRI-CGCM2.3.2, and (10)CCSM3. Reichler (2008) stated that the
average of global model outputs gives a better result than any individual model. Therefore, we can assume
that if we use models that are known to perform well in tropical areas, the composite (mean) value will
also give a better estimate for any parameter of interest compared to that of individual models.
2.4 Study of Climate Change and Its Impact in Indonesia
The study of climate change and its impacts can be done using two approaches, which are (1) bottom-
up, and (2) top-down (IPCC-TGICA, 2007) as shown in Figure 2.3. The bottom-up approach is based
on analysis of (historical) observational data and baseline data for both environment and socio-economic
9
parameters, while the top-down approach is based on GCM simulation results in correlation to scenarios
of changes in global environment and socio-economic conditions. The bottom-up approach can be used
to study climate changes that are currently happening or may occur in near future. On the other hand, the
top-down approach is the only way to obtain climate projections for the distant future.
Figure 2.3 Flowchart of the analysis of climate change and its impacts for two approaches, bottom-up
and top-down. (Source : IPCC, 2007b)
Studies of climate changes based on observational data for Indonesia have been conducted for example
by Manton et al. (2001), who tried to analyse the tendency of increasing extreme events in Asia-Pacific
region. In the study, they did not find a clear trend for temperature and rainfall changes in Indonesia.
However, it is noteworthy that one of the obstacles that they faced for Indonesia region is the very
limited data availability, particularly for temperature data. On the other hand, Boer and Faqih (2004)
analysed rainfall data from 210 stations and compare them with annual rainfall from 1931-1960 and 1961-
1990 period. This study reported that there is a trend of declining rainfall for some areas such as Java,
Lampung, South Sumatra, South Sulawesi, and Nusa Tenggara, but the trend suggests increasing rainfall
for areas such as Kalimantan and North Sulawesi. Meanwhile, Aldrian (2006) analysed the changes in
annual rainfall using observational data from 1950 to 1997 from some stations in Indonesia and found a
significant declining trend particularly in Kalimantan. These results show the difference in characteristics
of climate change that is occurring in Indonesia and indicate that it is difficult to obtain a comprehensive
picture of the overall trend.
10
The most recent information on changing trends in rainfall in Indonesia is obtained from Soetamto
(2009). In the presentation delivered during FGD discussion on February 24, 2009, an analysis of monthly
data for rainfall in Indonesia over 50 years (1951-2000) shows a linear trend. This result suggests that
the rainfall change trend is different for each month. However, the validity of this result is difficult to
evaluate as there are no information on the data source and its processing methodology. For this reason,
to meet the objective of this study, more data collection and analysis were needed.
The use of the top-down approach (using GCM data) for scientific studies on climate change in Indonesia
is also still very limited. Boer and Faqih (2004) analyse some GCM results including those of CSIRO Mark
2 (Australia), HadCM3 (England), and ECHAM-4 (Germany). They obtained a varied climate projection
particularly for rainfall. Santoso and Forner (2006) reviewed several existing studies and found that GCM
projections for rainfall in Indonesia have large uncertainties. A more comprehensive study that also covers
climate change risk analysis for agriculture is reported by Naylor, et al. (2007). Climate projection is done
using empirical (statistical) downscaling model for 20 GCM outputs and average rainfall for areas defined
in 8 regions. One of the important results of this study is the identification of climate change threat in
the form of increased probability of delay by up to 30 days in the onset of the rainy season in year 2050,
particularly in Java and Bali (Figure 2.4), which are the centre of rice agriculture in Indonesia. This threat
will greatly influence the rice planting activities, as these activities rely on water availability during the rainy
season. The applicability of the study done by Naylor et al. (2007) is quite limited because the analysis was
only performed for average rainfall in the defined areas for the year 2050.
Figure 2.4 Division of areas for study on climate change impact using IPCC AR-4 model by Naylor
et al. (2007) (left) and the analysis result of delay of the rainy season beginning (monsoon onset) for
Java island in the year 2050 (right). (Source: Naylor et al., 2007)
11
From the discussion above, we conclude that to compose the Roadmap using risk analysis approach
requires the availability of climate analysis and projection data, which is not available for all of Indonesia.
To understand the problem more clearly, we discussed it with a group of climate experts in Indonesia
in a closed (FGD) meeting on March 6-7, 2009, reaching a similar conclusion that, in Indonesia, the
well-evaluated climate projection output for the overall Indonesian area that can be adequately used for
sectoral analysis is not yet available in the literature. Therefore, with the existing obstacles, we tried to
conduct on our own the analysis and projection of climate change to meet the objectives of this study.
12
3
ANALYSIS AND
PROJECTION OF
CLIMATE CHANGE
IN INDONESIA
13
As explained earlier and following Boer and Faqih (2004), the analysis and projection of climate change
in Indonesia was carried out using two approaches: (1) historical data analysis (bottom-up) and (2)
GCM output analysis (top-down). In principle, the method is designed to be as simple as possible but
still to produce output that can be used by all sectors in estimating climate change risks. The data and
methodology that we used are explained below.
3.1 Data Availability
3.1.1 Observational Data
As detailed by Manton et al. (2001), climatology data with a good quality and continuity is not easy to
obtain for all of Indonesia. Considering that these analyses needed to be done immediately, we obtained
the data from sources on the internet that can be easily and immediately accessed. Here is the description
of the data available on the internet:
• GHCN (Global Historical Climatological Network) provides temperature and rainfall data for quite
a long period (1860s – now). However, data for the stations in Indonesia is generally only available
up through year 1975, except for a few stations. This data can be accessed online at http://www.
ncdc.noaa.gov/.
• GPCC (Global Precipitation Climatology Center) provides rainfall data in 0.5º x 0.5º grid form.
The data are available for a long time range of 1901-2007, even though some of the data are the
result of interpolation. This data can be accessed online at: ftp://ftp-anon.dwd.de/pub/data/
gpcc/html/fulldata_download.html.
Aside from the data mentioned above, we also had access to ASEAN Compendium of Climate Statistics
data (1975 – 1999). These datasets include rainfall and daily temperature data from all over Indonesia, but
the quality of data is limited, as there are many missing data. Even so, the data from several stations can
be processed as monthly data and can be used to complement GHCN data.
3.1.2 GCM Output Data
As previously explained, GCM output data can be accessed from the IPCC-AR4 data distribution centre.
In this regard, we obtained access to two websites that provide the data: (1) http://www.ipcc-data.org/
(IPCC Data Distribution Center) and (2) http://www-pcmdi.llnl.gov/(Program for Climate Model
Diagnosis and Inter-comparison). Because of various shortcomings, not all model output data have been
used for this study. Table 3.1 provides a simple description of GCM output data that we obtained from
these sites.
14
Table 3.1 Short description of GCM output data analysed in this study.
MODEL
Characteristics
NCCSM MPEH5 GFDLCM20 GFDLCM21 MIMR MRIGCM2 HADCM3
CCSR/
HC-
Developer NCAR, MPI, GFDL- GFDL- NIES/
MRI, Japan UKMO,
Institutes USA Germany NOAA, USA NOAA, USA JAMSTEC,
Japan
Japan
Resolution
2.8°x2.8° 1.8°x1.8° 2.5° x 2.0° 2.5° x 2.0° 2.8° x 2.8° 2.8° x 2.8° 2.75°x3.75°
(Lat x Long)
Scenario Simulation Year
1870-
20C3M 1860-2000 1861-2000 1861-2000 1850-2000 1850-2000 1860-1999
1999
2000-
SRESB1 2001-2200 2001-2300 2001-2300 2001-2300 1990-2300 2000-2199
2199
2000-
SRESA1B 2001-2200 2001-2300 2001-2300 2001-2300 2001-2300 2000-2199
2199
2000-
SRESA2 2001-2100 2001-2100 2001-2100 2001-2100 2001-2100 2000-2099
2099
From Table 3.1 it can be seen that GCM outputs resolution in general are very coarse; ranging around 2º, or
more than 200 km for tropical areas. Even so, for regional or macro-scale study, these data can already give
a description for climate projection. In several existing studies (Boer and Faqih, 2004; Santoso and Forner,
2006), the GCM data used only has coarse (global) resolution. On the other hand, although already using
empirical downscaling, Naylor et al. (2007) did not produce data with high resolution, but only areal average
data (or even provincial data) which are then corrected (calibrated) with observational data.
3.1.3 High-Resolution GCM Output Data
Aside from the downscaling technique, there is also global model output with high resolution. Such
experiments are conducted by MRI (Meteorological Research Institute) in Japan. Through BMKG,
we received MRI sample data with 20 km resolution (daily and monthly) which are used selectively in
this study. These data are part of the data used in an ongoing climate change study project (2009) in
BMKG.
15
3.2 Data Processing Method
3.2.1 Observational Data Processing
Observational data processing in general is meant to obtain climate change indicators by looking at the
mean and variance (standard deviation) of changes in the underlying trends. Overall, the analyses done
are basic statistical analysis and can be explained as follows:
(a) Moving average calculation of climate period (30 years) for each particular year interval. This
calculation is done for each month in a year.
(b) For each averaging period, the standard deviation is also calculated so that changes in climate
variability can be analysed.
(c) Empirical CDF (Cumulative Distribution Function) calculation is used to analyse the change in the
probability (frequency of occurrence) of high or low (rainfall) extreme.
(d) Projection of rainfall average value using polynomial trend analysis, which is different from previous
analyses that assumed a linear trend (e.g., Aldrian, 2006; Soetamto, 2009).
To analyse the changing spatial trends in rainfall, GPCC data is used because the GPCC data grid has a
size of 0.5º. In order to ensure its compatibility with data from station (GHCN), the GPCC data needs
to be evaluated first. We conducted the evaluation by choosing data samples from two stations i.e. Jakarta
and Ampenan (Lombok) as samples to compare the two sources of data and the result can be seen in
Figure 3.1 as a time series and in Figure 3.2 as a scatter plot. In this figures, the GPCC data value are
the result of interpolation to the coordinate of the station or its nearest point. It is noteworthy that the
station data are composite of GHCN data with other available data (ASEAN Compendium data for
Jakarta and data from BMKG for Ampenan).
Rainfall Data of JAKARTA (corr = 0.885) Rainfall Data of AMPENEN (corr = 0.876)
900 600
GPCC Data GPCC Data
GHCN Data GHCN Data
800
500
700
600 400
Rainfall (mm)
Rainfall (mm)
500
300
400
300 200
200
100
100
0 0
0 50 100 150 200 250 300 350 0 50 100 150 200 250 300 350
Jan1961 - Dec1990 Jan1961 - Dec1990
Figure 3.1 An example of time series comparison from GHCN (red) and GPCC (blue) for Jakarta
station (left) and Ampenan station (right).
16
Visually, we can see that there are differences between station data and GPCC data but both exhibit similar
trends. Figure 3.2 also shows that although the scattering of value is quite large, there is a significant linear
correlation, with a coefficient larger than 0.8. Hence, we can conclude that GPCC data can represent
observational data quite well. Nonetheless, it is to be noted that not all GPCC grid data are estimated
from the observation in the station because some data are purely derived from interpolation.
Under the assumption that GPCC data can represent observational data, most of our analysis has used
this set of data to obtain the current climate change trend, including the projection of mean value and its
standard deviation. As previously mentioned, the projection of mean value was done using polynomial
trend. This is because the changes in rainfall over time, in general, are not linear. For example, Figure 3.3
shows the moving average data for rainfall in January for 30 years in Jakarta station and its polynomial
trend analysis. In the figure, every data point represents the average for 30 years, although it is to be noted
that most of the data between 1940-1950 are empty and the average for 1995 only represent the data
between 1980-2007 (which is less than 30 years)
JAKARTA (corr = 0.885) AMPENEN (corr = 0.876)

800 600
700
500
600
400
500
GPCC Data
GPCC Data
400 300
300
200
200
100
100
0 0
0 100 200 300 400 500 600 700 800 900 0 100 200 300 400 500 600
GHCN Data GHCN Data
Figure 3.2 Comparison of two rainfall datasets from station (GHCN) and GPCC in scatter plot for
the same location as in Figure 5.
17
From Figure 3.3, it can be seen clearly that the changes of 30 year mean value follows a polynomial
pattern. Even so, we did not check for every data point and every month what degree of polynomial best
fits the data. For this reason, the best projection is found by calculating a weighted average value from the
available six estimations. This averaging method can be written mathematically as
P(tn ) = ∑ wi Pi (tn ), i = 1. Npoly

i
Here the weighted function for every polynomial estimation is determined by
wi =
∑ RMSE − RMSE i i
( Npoly − 1) × ∑ RMSE i
i
using RMSE (root mean square error) calculated from the difference between polynomial value and that
of observation for every data points, and Npoly = 6 is the number of polynomial function from first
order (linear) to sixth order.
data Projection linear quadratic cubic 4th degree 5th degree 6th degree
30-year mean rainfall (mm)
420
400
380
360
340
320
1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010
Year
20
Residuals (mm)
-20
1920 1930 1940 1950 1960 1970 1980 1990

Year
Figure 3.3 An example of 30 years moving average analysis using first order polynomial equation
(linear) to the sixth order (top) and its residual result (bottom)
18
3.2.2 GCM Data Processing
GCM data processing is basically done to find one model output or combination of several model outputs
that fits observational data. Based on the results of Reichler (2008) study, we assume that the ensemble
average of several model which are quite consistent for tropical area will be better representing climate
condition in Indonesia than an individual model. Figure 3.4 shows the spatial distribution of correlation
coefficient between baseline (1961-1990) composite annual rainfalls (average for every month) from
the output of one GCM (GFDLCM20) and GPCC data. From the figure we can see that the output of
GFDLCM20 model can represent the baseline climate pattern quite well for Java islands, Nusa Tenggara
islands, and some other areas in Sulawesi and Papua. But this is not the case for most of the area in
Kalimantan. Therefore, for each grid point the correlation coefficient for each model must be calculated
and only models with good enough correlation is averaged to obtain climate projection value. In this
case to obtain the projected value in every grid, we choose the best four models based on the correlation
coefficient value to be averaged. This approach is referred to as selective ensemble mean method. From
Figure 3.4, it can also be seen that using the ensemble mean, the output of mean rainfall gives the best fit
to observational data (GPCC data) although there is quite large in the GCM output values.
Figure 3.4 Distribution of composite baseline (1961-1990) correlation coefficient between the outputs
of rainfall data from GFDLCM20 model compared to GPCC (left), and an example of the result from
ensemble pattern for composite rainfall baseline for an area in Kalimantan (left). The full line with red
and black colour respectively represents the average value of ensemble and observation.
19
The processing of GCM output data for projection of spatial pattern of temperature is more difficult
to be done as there are no representative spatial observation data available because the global data is
only available on 2.5 degrees. For several observation points with the data available, we compare GCM
ensemble data with observational data. Figure 3.5 shows an example of comparison of GCM data with
observational data for surface temperature in two stations, Polonia-Medan and Penfui-Kupang. The
comparison for the two stations shows a quite similar pattern and order of magnitude between GCM
output data and observation, although it can also be clearly seen that there are bias or difference of average
and variance value. This is because of the coarse grid from GCM model so that the representation of
terrain condition is not quite good.
Considering the shortcomings of climatology baseline data for both rainfall and temperature, the
analysis of climate projection from GCM output data here will focus more on the trend of change in
climatological average value for rainfalls and temperature in the future relative to baseline period using
only model data (GCM). It is expected that this analysis will be still able to provide a description about
the existence of future climate change hazard.
Figure 3.5 Comparison of surface temperature data from GCM model output with the observed
surface temperature
20
4
RESULTS AND
DISCUSSIONS
21
4.1. Current Climate Change Trends in Indonesia
4.1.1 Rainfall Change
Based on the climate change detection criteria discussed above, the analysis for current climate change
trend has been viewed in terms of the changes for 30-year mean value. For example, Figure 4.1 shows
a moving average graphic of 30-year mean rainfall value for every 5 years from Jakarta station data. It
shows a quite significant increase of rainfall in January for the decade of the 1970s period compared to
the decade of 1900s period, with a difference in 30-year mean value that is close to 100 mm. Generally, it
is also shown that the rainfall in months around January to April is more sensitive to changes than other
months. Moreover, it can also be seen that rainfall in January tends to decline when it is close to year
2000s, while on the other hand rainfall in February tends to increase.
JAKARTA OBSERV. (30-year Annual Pattern) JAKARTA OBSERV. (30-year Average)

Lat = -6.17 , Lon = 106.82 Lat = -6.17 , Lon = 106.82
800 700
750 1891-1920 D
1911-1940 650 J
700 1931-1960 F
600
650 1951-1980
1971-2000 550
600
500
550
Average Rainfall (mm)
500 450
450 400
400 350
350 300
300
250
250
200
200
150 150
100 100
50 50
0 0
J A S O N D J F M A M J 1896 1906 1916 1926 1936 1946 1956 1966 1976 1986
Month Month
Figure 4.1 Changes in 30-year mean rainfall value for every month several period (left) and its moving
average graph for months in wet season D(ecember)-J(anuary)-F(ebruary) (right).
The above analysis shows that climate changes has occurred in Jakarta in certain level, as seen from the
changes in mean rainfall value from one 30-year period to the next. To find out the rainfall change trend
in all areas of Indonesia, a similar analysis is done for GPCC data in 0.5º grid size. As explained earlier,
we also try to project mean rainfall values based on polynomial trend analysis. Other examples of analytic
results are shown in Figure 4.2 and 4.3, while the detailed results are attached in Appendix A and B. In
Figure 4.2 can it be seen that, compared to the baseline period, there is a non-uniform change of mean
rainfall value for January. In Sumatera Island, most of the areas undergo an increase of mean rainfall
values ranging in 10-50 mm, while there are areas in which the average rainfall is increasing but in other
areas the average rainfall is declining. It is noteworthy that for example, in Java, Bali, and Nusa Tenggara,
there is a striking spatial variation where rainfall is declining over Lombok Island but an increasing trend
is observed over Bali.
22
Figure 4.2 An example of January rainfall trend analysis result from GPCC data: baseline 1961-1990
mean value (left) and the difference of current period 1980-2010 with baseline
(data available up to 2007)
The previously discussed results of trend analysis give an illustration of the current climate condition
compared with baseline period of 1961-1990. Furthermore, results of the polynomial extrapolation in
Figure 4.3 shows the trends for rainfall that may occur until year 2020 (see also Appendix B). Beside mean
value projection, using the same method, we also tried to produce projections of standard deviation that
can be used as an indicator of climate variability. The increasing value of the standard deviation value can
be interpreted as an increase in climate uncertainties as a result of the strength of climate variability in a
particular area. Both changes in mean value and standard deviation value contribute some to the potential
hazard of climate changes.
For the examples in January, it can be seen that almost all area in Sumatera, on average, still have the
potential to undergo rainfall increase until year 2020, with increasing uncertainties in the pattern of
rainfall, particularly for West Sumatera. The potential threat of changes in mean rainfall value and its
variability are projected to happen in 2015-2020 period. Decline of January rainfall tends to occur in parts
of Nusa Tenggara area. In Java Island, while some area undergo decline in rainfall, variability is in general
increasing although not to a great extent. In this case, the changes in mean rainfall value or its variance is
considered significant if it increases in the range of 25-50 mm or greater.
23
Figure 4.3 An example of January rainfall projection result expressed in the difference of 30-year
mean value (top) and standard deviation (bottom) for 1985-2015 period (left) and 1990-2020 period
(right) against baseline.
4.1.2 Surface Temperature Change
The projection of changes in surface temperature is difficult to develop with confidence because there
are no representative observational data (Manton et al., 2001; IPCC, 2007). Nonetheless, we tried to
analyse the temperature data for a time range that is quite long from several stations, in order to obtain an
illustration of surface temperature changes in Indonesia. The results of this analysis of temperature data
in Jakarta station are shown in Figure 4.4. The time series data in a glance shows a trend of significant
increase in temperature between the years in 1870s and 1980s. However, if we see the changes in 30 year
mean temperature value, we see that the changes during months in dry season June-July-August (JJA)
is greater than for months during wet season December-January-February (DJF). If we take its linear
trend, the changes in temperature for months in wet season is only around 0. 5º C but for months in dry
season it could reach 1.5º C during the 20th century. Temperature increase during the dry season months
in Jakarta is most likely influenced by local condition in the form of urban heat island. Urban heat island
effect is mostly generated by changes in land covering, while other contributors to the effect are heat
waste from industrial activities, transportation activities, and household activities.
24
30-year Annual Pattern
Time Series of Monthly Mean Temperature 30
JAKARTA OBSERV. Lat = -6.18 , Lon = 106.83
(Jan1866 - Feb2009)
29.5
29
29
28.5
Average Temperature ( C)
28
o
28
27.5
27
T ( C)
27
o
26.5 26
26
1879-1908
1899-1928
25.5 25 1919-1948
1939-1968
25 1959-1988
1979-2008
24.5 24
1860 1875 1890 1905 1920 1935 1950 1965 1980 1995 2010 J A S O N D J F M A M J
Time Month
30-year Average (DJF) 30-year Average (JJA)
29 29
D J
J J
F A
28.5 28.5
28 28
o
o
27.5 27.5
27 27
26.5 26.5
26 26
1904 1914 1924 1934 1944 1954 1964 1974 1984 1994 1904 1914 1924 1934 1944 1954 1964 1974 1984 1994
Month Month
Figure 4.4 Example of result from the analysis of temperature changes in Jakarta: (a) monthly
temperature time series data, (b) composite graph of 30-year temperature for each month (from July to
June), (c) Graph of 30-year moving average for December, January, February, and (d) similar to (c) for
June, July, August.
4.1.3 Probability Change for Extreme Weather and Climate
The analysis for extreme weather events actually needs more complete data. The analysis of extreme
rainfall events, for example, needs at least daily rainfall data, or even hourly data. Nonetheless, more
information can be retrieved from monthly rainfall data from the observation in the stations. In Figure
4.5 an empirical Cumulative Distribution Function (CDF) graph is shown from different 30-year periods
just as seen in Figure 4.1. From the figure it can be seen that for December-January-February months
(DJF), the chance of monthly rainfall value of 500 mm increased by 10% (from 10% to 20%)in 1961-
1990 period compared to 1901-1930 period and 1931-1960 period. On the other hand, the change in
probability distribution for rainfall in the months of March-April-May (MAM) is not quite clear. It needs
to be noted that rainfall probability distribution for year 1991-1997 is difficult to interpret as there are not
25
enough data to form a normal graph. The correlation of high and extreme rainfall probability is not quite
clear. However, the average rainfall in Jakarta is around 200-300 mm/month during DJF, and rainfall
value of 500 mm/month is almost twice the average. If we assume the number of rainy days is constant,
then it can be estimated that there is an increase of daily rainfall intensity as well.
Empirical CDF (DJF) Empirical CDF (MAM)

1 1
0.9 0.9
0.8 0.8
0.7 0.7
Cumulative Probability
Cumulative Probability
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
1901-1930 1901-1930
1931-1960 1931-1960
0.1 0.1
1961-1990 1961-1990
1991-1997 1991-1997
0 0
0 100 200 300 400 500 600 700 800 900 0 100 200 300 400 500 600 700 800 900
Rainfall (mm/month) Rainfall (mm/month)
Figure 4.5 Empirical Cumulative Distribution Function (CDF) graph, from Jakarta monthly rainfall
data for some 30-year periods.
In projection of extreme weather events probability, GCM output data with resolution of 20 km from MRI
can provide additional information. Nonetheless, it needs to be noted that the data require a calibration
process. We selectively tried to analyse the distribution of rain probability from the MRI-GCM model
output for the following years: 2015, 2020, 2025, and 2030. Figure 4.6 shows a graph of the cumulative
probability for rainfall value for Java-Bali area (land area only). This figure shows that the rainfall value of
20 mm/day is already located within 5% of the upper limit.
26
1
0.9
0.8
Distribusi Peluang Kumulatif

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 10 20 30 40 50 60 70 80 90 100
Curah Hujan (mm/hari)
Figure 4.6 Similar to figure 4.5 but for daily rainfall from MRI-GCM model output data for Java-Bali
area. The dashed red line show the value of 1 mm (vertical)
and the 0.95 probability function value (horizontal)
To see the change in probability of extreme rainfall, a distribution map of probability of rainfall greater
than 20 mm is shown in Figure 4.7. According to MRI-GCM simulation result, it can be seen that
areas with probability of extreme rain in Java Island are more scattered in 2015 than in 2020. However,
the chances of extreme rainfall events around Jakarta area will be greater in 2020 than in 2015. This
projection is consistent with the analysis result using GPCC data (Appendix B2) where the trend change
from 1981-2010 period to1990-2020 period shows a higher variability compared to the trend change
from 1981-2010 to 1985 -2015 for Jakarta area.
27
Figure 4.7 Distribution of high daily rainfall probability (>20 mm/day based on GCM model output)
in Java-Bali area for year 2015 (left) and 2020 (right).
4.2 Climate Change in 2030s Period and 2080 Period
A climate projection for distant future can only be obtained from GCM output analysis. Because GCM
output is a 4-dimensional data (latitude, longitude, time, scenario), it is very difficult to present all data.
To present the information in a simpler way, we only review the change in temperature and rainfall for
2020-2050 (2030s) period and for 2070-2100 (2080s) period against baseline period (1961-1990). Then
every region is divided into several areas and GCM output data is averaged for each area. For example, as
seen in Figure 4.8, Java-Bali region is divided into 4 areas: (1) Western Part, (2) Central Part, (3) Eastern
Part, and (4) Bali.
4.2.1 Projection of Surface Temperature Change
The projection of surface temperature for the four areas in Java-Bali region can be seen in Figure 4.9.
From the picture it can be seen that the projection of mean temperature change generally does not
very much from one area to the other. Moreover, temperature change also follows global trend with an
increase that is almost linear.
28
Figure 4.8 The division of Java-Bali region into four areas for the analysis of climate projection.
Different scenarios give significantly different temperature in the projection for 2080s period, while the
average temperature increase in 2030s period does not show a substantial change. As expected, the A2
scenario with the highest CO2 concentration in 2100 gives the highest increase, around 3º C. On the
other hand, the B1 scenario only gives an increase of 2º C, and A1B scenario gives a value in-between
these two extremes. This result shows that in the modelling, the increase of averaged surface temperature
indeed follows a relatively simple mechanism where the addition of heat is determined by the balance of
infrared radiation, which is dictated by CO2 concentration The graph for temperature increase for other
areas can be seen in Appendix D1-D7.
29
Figure 4.9 Projection of temperature increase of 2030s period (left) and 2080s period (right) for
scenario B1, A1B, and A2 (bottom to top) over Java-Bali region (see Figure 4.8) .
30
4.2.2 Projection of Rainfall Change
The projection of changes of rainfall pattern over Java-Bali region in general is more varied than the pattern
of temperature increase, which has linear tendency. In principle, the potential danger we need to be wary
of is the tendency of increasing rainfall in wet months and the decline of rainfall in transitional months
compared to the baseline condition (1961-1990). According to GCM output analysis, this condition will
be exacerbated near 2080s period. In regard to the decline of rainfall in dry months, Western Java area is
the area with higher potential hazard compared to the other areas. It should also be noted that there are
a large variation for January rainfall projection for all three SRES scenarios.
Although it is not as simple as temperature pattern, the influence of the scenario used also seems to be
proportional to the CO2 concentration, where scenario A2 with the highest concentration also produces
the largest magnitude of rainfall change. This GCM output analysis also gives an illustration of changes
in the monthly pattern so it will be known that, for example, Eastern Java and Bali will undergo rainfall
increase in February compared to the Western Java area in the 2030s period.
4.2.3 Projection of Probability Change for Extreme Weather and Climate
The analysis for projection of change in extreme weather and climate probability is yet to be done so in
this stage the result is not presentable. Because of the limited data availability, there is not much that we
can do to identify potential hazards due to extreme weather. Even so, the hazard of extreme climate such
as the occurrence of consecutive dry years become one of many things that we should be wary of. This
issue will be best discussed and reviewed in future revision of this study.
31
Figure 4.10 Projection of rainfall increase of 2030s period (left) and 2080s period (right) for scenario
B1, A1B, and A2 (bottom to top) over Java-Bali region
32
5
GENERAL
RECOMMENDATION
FOR SECTORAL
ADAPTATION
33
5.1 Description of Climate Change for Each Region
As the basis for the making of Roadmap, we tried to formulate the scientific basis study result in a simple
form that is highly informative. However, this is not easily done both because the climate problem
in Indonesia is very complex and also because of numerous limitations in the availability of climate
projection data. Below we present concisely the general description of climate change projections and
its probable hazard for each region based on polynomial trend analysis (for rainfall until year 2020) and
GCM output analysis (for 2030s and 2080s period), while the detailed version of these results can be seen
in Appendix B, D and E.
5.1.1 Climate Change in Java-Bali Area
Climate changes in Java-Bali area has been discussed in the examples of climate projection analysis above.
Based on polynomial trend analysis, in short, we note that relative to the current condition (averaged
over 1981-2007), rainfall change between 2010 until 2015 is not a significant departure form recent
climatological mean. However, the average rainfall for year 2010 until 2020 shows a decline in rainfall
for January and an increase of rainfall for December and February in most of the areas in West Java and
East Java. Only in some areas on northern coast of Central Java there are increases of rainfall for both
January and February. As previously mentioned, rainfall changes are considered as significant when it is
in the range or exceed the range of 25-50 mm.
Projections using GCM models show that the increase of average surface temperature in 2020-2050
period is about 1º C for all scenarios relative to baseline. In the 2070-2100 period, the temperature increase
is around 2º C for scenario B1, 2.5º C for A1B and 3º C for A2. The pattern of rainfall change shows
a general tendency that rainfall in dry months (June-July-August) tends to decline while the increase of
rainfall happens during wet months (December-January-February), especially on 2080s period for A1B
and A2 scenario. It is also to be noticed that the January rainfall in A2 and A1B scenario show a pattern
of larger spatial variation, with a tendency of increasing rainfall in February.
5.1.2 Climate Change in Sumatera Area
Rainfall in some areas of Sumatera currently has a tendency to increase relative to baseline and is projected
to keep increasing until 2020. The potential hazard lies significantly in the increase of mean rainfall and
its variability in West and North Sumatera in October. Potential hazard of increasing rainfall variability
also lies more specifically in April for West Sumatera.
GCM model projections show that the increase of average surface temperature in 2020-2050 period is
around 1º C for all scenarios relative to baseline, but it varies more from one month to the other between
34
0-2º C. During 2070-2100 period, the temperature increase is around 2º C for B1 scenario, 2.5º C for
A1B, and 3º C for A2, but the largest increase can reach 4º C. Rainfall change pattern shows difference for
Southern Sumatera compared to other parts of Sumatera. The general tendency for Central to Northern
Sumatera area can be regarded as the contrast to the pattern occurring in Java-Bali and Southern Sumatera,
particularly the pattern of increasing rainfall for January.
5.1.3 Climate Change in Kalimantan Area
The estimated increase of rainfall in Eastern Kalimantan is based on trend analysis and occurs particularly
around January when the variability also increases during 2010-2020.
General changes in temperature in Kalimantan area follow the global trend, which is 1º C for the 2030s
period and 1.5º C - 3º C for the 2080s period, with a greater variation compared to Java but less than the
variation in Sumatera. The estimated change of rainfall from GCM output analysis does not show a clear
pattern but rainfall increase is more dominant for 2080s period for A1B and A2 scenarios. However,
rainfall is projected to decline during the months of January-February-March in North Western
Kalimantan area.
5.1.4 Climate Change in Sulawesi Area
Potential hazard of changes in rainfall for Sulawesi area during 2010-2020 lies in the increase of rainfall
in December and January, particularly for Northern Sulawesi area.
Temperature change in general follows global trend which is around 0.8º C in the 2030s period and 1.5º
C - 3º C for 2080s period, with a quite small variation. Changes in rainfall show a highly varied pattern
with a striking difference between Southern and Northern Sulawesi Areas, particularly for A1B scenario
in the 2080s period.
5.1.5 Climate Change in Nusatenggara Area
Projection using GPCC data shows a striking increase in potential danger of climate variability in Western
Nusatenggara for November months during 2010-2020 period. Meanwhile, January rainfall tends to
decline significantly.
Based on GCCM output analysis, change of temperature in general follows a global trend which is
increasing by around 0.8º C during 2030s period and 1.5º C - 3º C during 2080s period with a quite large
variation between wet months and dry months. Change of rainfall shows a highly varied pattern with
more dominant tendency to decline slightly in the 2080s period for months between March to September.
The increase of rainfall in October appears significantly in almost all scenarios and periods.
35
5.1.6 Climate Change in Moluccas Area
Change in rainfall in Moluccas area has a potential to undergo increase and decline during 2010-2020
period in different months. The potential of increased mean rainfall value and its variability mainly lies in
June around Seram Island, which contrasts with the tendency of rainfall increase for Halmahera Island
that is projected to occur in January.
Temperature change in general follows global trend of GCM output but is slightly smaller than 1º C-3º
C between 2030 period until 2080s. Changes in rainfall are highly varied with a tendency of declining
rainfall in 2030s but then increase happens in 2080s particularly in Northern area. Rainfall projection is
quite varied among scenarios, showing high model uncertainties for Moluccas area.
5.1.7 Climate Change in Papua Area
Change in rainfall in Papua area has the potential for declining tendency during 2010-2020 period, except
for June and July months when increase occurs. Meanwhile, the potential of increasing climate variability
lies particularly in July.
In general, changes in temperature follow the global trend which is 1 º C-3 º C between 2030s period until
2080s period, with a variation that is quite small except for Southern Papua. Furthermore, GCM output
data analysis shows that the changes in rainfall tend to increase significantly, except for Southern Papua,
particularly for 2080s period and show a consistent pattern for all scenarios.
5.2 Additional Notes
5.2.1 Long Term Climate Variability
Our understanding of regional climate change is very limited in a number of perspectives, including the
past and the present, and even less sure about the future. Even so, in facing climate change issue, the main
problem we are facing is the increase of Earth surface temperature caused by the increase in greenhouse
gases concentrations. In general, the existence of global temperature increase during the post-industrial
era has been acknowledged by the worlds’ climate experts. Using science and technology as far as possible,
the increase of global temperature has been confirmed by scientists using GCMs (Global Circulation
Models) for 20th century condition. However, scientists are also aware that GCM success in simulating
current climate condition does not guarantee the validity of its climate projection for the next 100 years.
Aware of these shortcomings, institutions that develop climate models also keep improving the structure
and output of their models.
In taking action to respond to future climate changes, aside from prediction of climate change based on
model output, other things that need to be considered include the long-term climate variability (inter-
36
decadal) such as PDO (Pacific Decadal Oscillation), which is related to the sea surface temperature
variation in Northern Pacific and Central Pacific (Mantua, 1997), and AMO (Atlantic Multidecadal
Oscillation) which is related to the thermohaline (deep sea) circulation changes in the Atlantic Ocean
(Soon, 2009). The existence of long-term climate variability has caused the attribution of global warming
to climate change be debated more fiercely among world’s climate scientists. The IPCC report (2007,
Chapter 6) discussed the debate on long-term climate variability, but concluded that the current and
historical climate change cannot be separated from anthropogenic influence.
The effect of long-term variability also needs to be considered in the analysis of projections of future
climate impacts so that not all climate pattern shifts will easily be attributed to global warming. In Figure
5.1 it is shown PDO index time series and rainfall data taken from rainfall observation data from Ampenan
Station (Lombok Island) after smoothing out for 20-year variation. The PDO index data can be obtained
from the site http://jisao.washington.edu/pdo/PDO.latest . In this analysis the rainfall data is converted
to rainfall index and by calculating the anomaly and dividing it by a scale factor. From Figure 5.1 we see that
there is a correspondence in temporal variation pattern between PDO index variations with the rainfall in
Lombok Island. This shows a strong possibility of the connection between current climate changes and
long term variability that affects inter-decadal (30 years) rainfall statistics. The linear correlation tendency
is also visible when the rainfall index is shifted 14 years backward (Hadi et al., 2009)
1
Rainfall/PDO Index
-1
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010
Year
Figure 5.1 Time-series graph of Pacific Decadal Oscillation (PDO, red) and rainfall index (blue) from
Ampenan Station (Lombok Island) after smoothing with 20-year period.
The latest study by Soon (2009) shows that the long term atmosphere variability is affected by the
dynamical processes in the Atlantic Ocean through a mechanism known as meridional overturning
circulation (MOC), which is generated by solar radiation variability. Figure 5.2 shows a time series graph
of sea surface temperature in Atlantic and total solar irradiance that show a quite clear similarity of
long-term variation pattern. In this note we first want to emphasize the importance of strengthening
37
the academic basis of study on climate dynamics in Indonesia so that the climate change behaviour can
be understood more comprehensively. Secondly, the potential hazard of climate change from long-term
climate variability is not of lesser importance to be considered compared to climate change due to global
warming.
Figure 5.2 Time-series graphic of annual sea surface temperature anomaly in Atlantic (dashed line) and
total solar irradiance (red line). (Source: Soon, 2009)
5.2.2 Climate Projection Uncertainties
As mentioned in the beginning of the study, climate projection always contains uncertainties and the
biggest challenge is to quantify the uncertainties in order to improve the usefulness of these projections
for decision-making. In making climate projections from GCM data, there are at least three sources of
uncertainties that needs to be considered, those are: (1) uncertainties in the greenhouse gases emission
scenarios, (2) uncertainties about global climate sensitivity to greenhouse gases concentration (GCM
model choice), and (3) uncertainties about the response of regional climate systems towards global
warming (downscaling model).
As an illustration, this study considers three emission scenarios (B1, A1B, A2) and seven GCM output
without downscaling. In combination, this creates 3×7 = 21 possibilities of projected future climate
change. For a more comprehensive analysis, hundreds of scenario combinations may be used to produce
38
hundreds of possibilities from which the statistics can be studied. In the present work, we only presented
the best estimate in the form of selective ensemble mean value with uncertainty information taken only
from three SRES scenarios.
39
40
6
CLOSING REMARK
41
In supporting the decision-making process of the Roadmap, we have tried to conduct our study on
future climate change in Indonesia as comprehensively as possible. Even so, considering the numerous
limitations on data and methodology that we confront, the results presented here must be considered as
provisional and far from meeting the national needs for more accurate climate change information. For
this reason, in the future a more coordinated effort to strengthen the climate change scientific basis is
needed by empowering relevant institutions in a more optimal way. Therefore, the Roadmap is not only
needed to guide sectoral adaptation action but must also be used as a foundation for strengthening the
scientific basis.
Currently there are numerous studies on the impacts and the risks of climate change conducted without
adequate scientific basis study. For climate change issue not to become a mere political commodity, in the
future we see the need to compose a book that is equivalent with Climate Change 2007 The Physical
Science Basis, or say Climate Change in Indonesia 20xx: The Science Basis that compiles the best
contribution from all competent institutions and individuals in discussing climate change in Indonesia.
42
REFERENCES
Aldrian, E. and R.D. Susanto, 2003 : Identification of three dominant rainfall regions within Indonesia
and their relationship to sea surface temperature, Int. J. Climatol., 23, 1453-1464
Aldrian, E., 2006 : Decreasing trends in annual rainfalls over Indonesia: A Threat to national water
resource?, J. Meteorologi dan Geofisika (in Press version).
Annamalai, H., K. Hamilton, and K.R. Sperber, 2006 : South Asian Summer Monsoon and Its Relationship
with ENSO in the IPCC AR4 simulations, J. Climate (in Press)
Australian Government, Department of the Environment and Heritage Australian Greenhouse Office,
2005 : Climate Change Risk and Vulnerability, (http://www.greenhouse.gov.au/ )
Boer, R. and Faqih, A. 2004 : Current and Future Rainfall Variability in Indonesia, Paper submitted to
Assessments of Impacts and Adaptations to Climate Change (AIACC), ( http://www.enfor.com.ph/
publications.html )
Hadi, T.W., Halil, and Ibnu Sofian, 2009 : Recent Trend in The Interannual and Interdecadal Rainfall
Variation over Lombok Island, Indonesia, In Preparation (to be submitted to SOLA in July )
IPCC, 2007: Climate Change 2007 : The Physical Science Basis. Contribution of Working Group I to the
Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin,
M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. Cambridge University
Press, Cambridge, United Kingdom and New York, NY, USA, 996 pp.
IPCC-Task Group on Data and Scenario Support for Impacts and Climate Analysis (TGCIA), 2007
: General Guidelines on The Use of Scenario Data for Climate Impact and Adaptation Assessement,
Version 2, pp. 66
Joseph, R., and S. Nigam, 2006 : ENSO Evolution and Teleconnections in IPCC’s Twentieth-Century
Climate Simulations: Realistic Representation?, J. Climate, 19, 4360-4377
Kim, J., and T. Reichler, 2008 : Regional Performance of the IPCC-AR4 Models in Simulating Present-
Day Mean Climate, EGU General Assembly, Vienna, Austria, April 13-18, (http://www.inscc.utah.edu/
~reichler/talks/tjr_talks.shtml)
Manton, M.J., P.M. Della-Marta, M.R. Haylock, K.J. Hennessy, N. Nicholls, L.E. Chambers, D.A. Collins,
G. Daw, D. Gunawan, K. Inape, H. Isobe, T.S. Kestin, P. Lefale, C.H. Leyu, T. Lwin, L. Maitrepierre,
N. Ouprasitwong, C.M. Page, J. Pahalad, N. Plummer, M.J. Salinger, R. Suppiah, V.L. Tran, B. Trewin,
I. Tibig, and D. Yee, 2001, Trends in extreme daily rainfall and temperature in Southeast Asia and the
Southeast Pacific : 1961-1998, Int. J. Climatol., 21, 269-284.
43
Mantua, N.J. and S.R. Hare, Y. Zhang, J.M. Wallace, and R.C. Francis, 1997 : A Pacific interdecadal climate
oscillation with impacts on salmon production. Bull. Amer. Meteor. Soc., 78, 1069-1079
Meehl, G.A., F. Zwiers, J. Evans, T. Knutson, L. Mearns, and P. Whetton, 2000 : Trends in extreme
weather and climate events : Issues related to modeling extremes in projections of future clmate change,
Bull. Amer. Met. Soc., 81(3), 413-416
Naylor, R.L., D.S. Battisti, D.J. Vimont, W.P. Falcon, and M.B. Burke, 2007 : Assessing risk of climate
variability and climate change for Indonesian rice agriculture, PNAS, 104(19), 7752-7757
Pielke, R.A., Jr., 2003 : What is climate change? Policy consequences of differing
political and scientific definitions, pp. 7, http://www.climateadaptation.net/
docs/papers/pielke.pdf
Santoso, H., and C. Forner, 2006 : Climate change projections for Indonesia, Background document for the
Southeast Asia kick-off meeting of the project Tropical Forests and Climate Change Adaptation (“TroFCCA”), Bogor,
May 29-30
Schneider, S., 2002 : Can we estimate the likelihood of climatic changes at 2100?, Climatic Change, 52,
441-451
Soetamto, 2009 : Analisa Perubahan Hujan di Indonesia, Materi FGD Sektor Air dan Pengelolaan
Limbah, Bappenas, 24 Februari
Soon, W.-H, 2009 : Solar Arctic-Mediated Climate Variation on Multidecadal to Centennial Timescales:
Empirical Evidence, Mechanistic Explanation, and Testable Consequences, Preprint of publication to appear
in Physical Geography (February 12)
Smith, I., 2009 : Current Issues with Climate Change Projections, AusAID Workshop 2009 on High
Resolution Climate Modelling of Climate Change over the Indonesian Region, CSIRO Marine and
Atmospheric Research, Aspendale, 18-29 May
Wilby, R.L., S. Charles, L.O. Mearns, P. Whetton, E. Zorito, B. Timbal, 2004 : Guidelines for use
of climate scenarios developed from statistical downscaling methods, IPCC Task Group on Data
and Scenario Support for Impacts and Climate Analysis (TGCIA), (http://ipcc-ddc.cru.uea.ac.uk/
guidelines/StatDown_Guide.pdf)
44
APPENDIX A : RAINFALL BASELINE BASED ON GPCC DATA
45
46
APPENDIX B1 : RAINFALL CHANGE PROJECTION 2010-2015
APPENDIX F : STATION RAINFALL GRAPH
47
48
49
50
APPENDIX B2 : RAINFALL CHANGE PROJECTION 2015-2020
51
52
53
APPENDIX C : AREA DIVISION
54
55
APPENDIX D1 : TEMPERATURE PROJECTION FOR 2030s AND 2080s IN JAVA BALI
56
APPENDIX D2 : TEMPERATURE PROJECTION FOR 2030s AND 2080s IN KALIMANTAN
57
APPENDIX D3 : TEMPERATURE PROJECTION FOR 2030s AND 2080s IN MALUKU
58
APPENDIX D4 : TEMPERATURE PROJECTION FOR 2030s AND 2080s IN NUSA TENGGARA
59
APPENDIX D5 : TEMPERATURE PROJECTION FOR 2030s AND 2080s IN PAPUA
60
APPENDIX D6 : TEMPERATURE PROJECTION FOR 2030s AND 2080s IN SULAWESI
61
APPENDIX D7 : TEMPERATURE PROJECTION FOR 2030s AND 2080s IN SUMATERA
62
APPENDIX E1 : RAINFALL PROJECTION FOR 2030s AND 2080s IN JAVA-BALI
63
APPENDIX E2 : RAINFALL PROJECTION FOR 2030s AND 2080s IN KALIMANTAN
64
APPENDIX E3 : RAINFALL PROJECTION FOR 2030s AND 2080s IN MALUKU
65
APPENDIX E4 : RAINFALL PROJECTION FOR 2030s AND 2080s IN NUSA TENGGARA
66
APPENDIX E1 : RAINFALL PROJECTION FOR 2030s AND 2080s IN PAPUA
67
APPENDIX E1 : RAINFALL PROJECTION FOR 2030s AND 2080s IN SULAWESI
68
APPENDIX E1 : RAINFALL PROJECTION FOR 2030s AND 2080s IN SUMATERA
69
APPENDIX F : STATION RAINFALL GRAPH
BLANG BINTANG BANDA (30-year Annual Pattern) PIRU (30-year Annual Pattern)
Lat = 5.52 , Lon = 95.43 Lat = -3.07 , Lon = 128.18
800 800
750 1891-1920 750 1891-1920
1911-1940 1911-1940
700 1931-1960 700 1931-1960
650 1951-1980 650 1951-1980
1971-2000 1971-2000
600 600
550

550
500 500
450 450
400 400
350 350
300 300
250 250
200 200
150 150
100 100
50 50
0 0
J A S O N D J F M A M J J A S O N D J F M A M J
Month Month
PAYAKUMBUH (30-year Annual Pattern) PANGKAL PINANG (30-year Annual Pattern)

Lat = -0.22 , Lon = 100.62 Lat = -2.17 , Lon = 106.13
800 800
750 1891-1920 1891-1920
1911-1940 750
1911-1940
700 1931-1960 700 1931-1960
650 1951-1980 1951-1980
650
1971-2000 1971-2000
600 600
550
550
500 500
450 450
400 400
350 350
300 300
250 250
200 200
150 150
100 100
50 50
0 0
Month Month
BELINYU (30-year Annual Pattern) SINGKAWANG (30-year Annual Pattern)

Lat = -1.65 , Lon = 105.77 Lat = 0.92 , Lon = 109
800 800
750 1891-1920 1891-1920
1911-1940
750
1911-1940
700 1931-1960 700 1931-1960
650 1951-1980 650 1951-1980
1971-2000 1971-2000
600 600
550
550
500 500
450 450
400 400
350 350
300 300
250 250
200 200
150 150
100 100
50 50
0 0
Month Month
70
JAKARTA OBSERV. (30-year Annual Pattern) BOGOREMPANG (30-year Annual Pattern)
Lat = -6.17 , Lon = 106.82 Lat = -6.6 , Lon = 106.78
800 800
750 1891-1920 750 1891-1920
1911-1940 1911-1940
700 1931-1960 700 1931-1960
650 1951-1980 650 1951-1980
1971-2000 1971-2000
600 600
550
550

500 500
450 450
400 400
350 350
300 300
250 250
200 200
150 150
100 100
50
50
0
J A S O N D J F M A M J 0
J A S O N D J F M A M J
Month
Month
CIANJUR (30-year Annual Pattern) JATIWANGI (30-year Annual Pattern)

Lat = -6.82 , Lon = 107.45 Lat = -6.75 , Lon = 108.27
800 800
750 1891-1920 750 1891-1920
1911-1940 1911-1940
700 1931-1960 700 1931-1960
650 1951-1980 650 1951-1980
1971-2000 1971-2000
600 600
550
550
500 500
450 450
400 400
350 350
300 300
250 250
200 200
150 150
100 100
50 50
0 0
Month Month
BANJARNEGARA (30-year Annual Pattern) SIKAYU (30-year Annual Pattern)

Lat = -7.4 , Lon = 109.7 Lat = -7.68 , Lon = 109.48
800 800
1891-1920 750 1891-1920
750
1911-1940 1911-1940
700 1931-1960 700 1931-1960
650 1951-1980 650 1951-1980
1971-2000 1971-2000
600 600
550
550
500 500
450 450
400 400
350 350
300 300
250 250
200 200
150 150
100 100
50 50
0 0
Month Month
71
KARANGASEM (30-year Annual Pattern)
Lat = -7 , Lon = 111.1 TUGUREJO (30-year Annual Pattern)
800 Lat = -6.98 , Lon = 110.35
1891-1920 800
750 1891-1920
1911-1940 750
700 1931-1960 1911-1940
700 1931-1960
650 1951-1980
1971-2000 650 1951-1980
600 1971-2000
600
550
550

500
500
450
450
400
400
350
350
300
300
250
250
200
200
150
150
100
100
50
50
0
Month
Month
TANJUNGTIRTO (30-year Annual Pattern) NGLANGON (30-year Annual Pattern)

Lat = -7.8 , Lon = 110.47 Lat = -7.15 , Lon = 111.15
800 800
750 1891-1920 1891-1920
750
1911-1940 1911-1940
700 1931-1960 700 1931-1960
650 1951-1980 650 1951-1980
1971-2000 1971-2000
600 600
550
550
500 500
450 450
400 400
350
350
300
300
250
250
200
200
150
150
100
100
50
50
0
Month J A S O N D J F M A M J
Month
SRUNI (30-year Annual Pattern)

PANDANLARAS (30-year Annual Pattern)
Lat = -7.4 , Lon = 112.72
Lat = -7.9 , Lon = 113.48
800 800
750 1891-1920 1891-1920
1911-1940 750
700 1911-1940
1931-1960 700 1931-1960
650 1951-1980 1951-1980
650
1971-2000 1971-2000
600 600
550
550
500 500
450 450
400 400
350 350
300 300
250 250
200 200
150 150
100 100
50 50
0 0
Month Month
72
NEGARA (30-year Annual Pattern)
KABAT (30-year Annual Pattern)
Lat = -8.33 , Lon = 114.65
Lat = -8.27 , Lon = 114.32 800
800
750 1891-1920
750 1891-1920
1911-1940
1911-1940 700
700 1931-1960
1931-1960
650 1951-1980
650 1951-1980
1971-2000
1971-2000 600
600
550

550
500
500
450
450
400
400
350
350
300
300
250
250
200
200
150
150
100
100
50
50
0
0 J A S O N D J F M A M J
Month
Month
BENTENG (30-year Annual Pattern) REMBIGA AMPENAN (30-year Annual Pattern)

Lat = -6.12 , Lon = 120.47 Lat = -8.53 , Lon = 116.07
800 500
1891-1920 1891-1920
750 1911-1940
1911-1940 450
700 1931-1960 1931-1960
650 1951-1980 400 1951-1980
1971-2000 1971-2000
600
350
550
500
300
450
400 250
350
300
200
250 150
200
150 100
100
50
50
0 0
Month Month
LABUHA (30-year Annual Pattern) PATTIMURA/AMBON (30-year Annual Pattern)

Lat = -0.65 , Lon = 127.47 Lat = -3.7 , Lon = 128.1
800 800
750 1891-1920 1891-1920
750
1911-1940 1911-1940
700 1931-1960 700 1931-1960
650 1951-1980 650 1951-1980
1971-2000 1971-2000
600 600
550
550
500 500
450 450
400 400
350 350
300 300
250 250
200 200
150
150
100
100
50
50
0
Month
Month
73

Analysis and Projection of Temperature and Rainfall 20110217130224 0

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Analysis and Projection of Temperature and Rainfall 20110217130224 0

Uploaded by

Copyright:

Available Formats

Scientific Basis:

Analysis and Projection of Temperature and Rainfall

Scientific Basis: Analysis and Projection of Temperature and Rainfall

Prof. Armida S. Alisjahbana, Minister of National Development Planning/Head of Bappenas

Edi Effendi Tedjakusuma, Director of Environmental Affairs, Bappenas

Irving Mintzer, Syamsidar Thamrin, Heiner von Luepke, Dieter Brulez

Coordinating Authors for Adaptation: Djoko Santoso Abi Suroso

Scientific Basis: Analysis and Projection of Temperature and Rainfall Report

Author: Tri Wahyu Hadi

Technical Supporting Team

Steering Committee (SC)

Deputy of International Cooperation, Coordinating Ministry for Economy; Secretary of Minister,

National Development Planning Agency

Agency for Meteorology, Climatology and Geophysics

Edvin Aldrian, Dodo Gunawan, Nurhayati, Soetamto, Yunus S, Sunaryo

National Institute of Aeuronatics and Space

Agus Hidayat, Halimurrahman, Bambang Siswanto, Erna Sri A, Husni Nasution

Research and Implementatiton of Technology Board

Eddy Supriyono, Fadli Syamsuddin, Alvini, Edie P

National Coordinating Agency for Survey and Mapping

Suwahyono, Habib Subagio, Agus Santoso

Minister for National Development Planning/

Prof. Armida S. Alisjahbana

To be a part of the solution to global climate change, the government

With a unique set of geographical location, among countries on the

Deputy Minister for Natural Resources and Environment

Remarks from Minister of National Development Planning/Head of Bappenas v

TABLE OF CONTENTS viii

LIST OF FIGURES xii

LIST OF ABBREVIATION xiv

1.1 Background and Objectives 2

1.2 Scope of Discussions and Contents of This Report 3

2.1 Global Climate System and Definition of Climate Change 6

2.2 Global Climate Change Scenario and Projection 7

2.3 Global Circulation Model (GCM) Performance in Tropical Area 9

2.4 Study of Climate Change and Its Impact in Indonesia 9

3. ANALYSIS AND PROJECTION OF CLIMATE CHANGE IN INDONESIA 13

3.1 Data Availability 14

3.1.1 Observational Data 14

3.1.2 GCM Output Data 14

3.1.3 High-Resolution GCM Output Data 15

3.2 Data Processing Method 16

3.2.1 Observational Data Processing 16

3.2.2 GCM Data Processing 19

4.1. Current Climate Change Trends in Indonesia 22

4.1.1 Rainfall Change 22

4.1.2 Surface Temperature Change 24

4.1.3 Probability Change for Extreme Weather and Climate 25

4.2 Climate Change in 2030s Period and 2080 Period 28

4.2.1 Projection of Surface Temperature Change 28

4.2.2 Projection of Rainfall Change 31

4.2.3 Projection of Probability Change for Extreme Weather and Climate 31

5. GENERAL RECOMMENDATION FOR SECTORAL ADAPTATION 33

5.1 Description of Climate Change for Each Region 34

5.1.1 Climate Change in Java-Bali Area 34

5.1.2 Climate Change in Sumatera Area 34

5.1.3 Climate Change in Kalimantan Area 35

5.1.4 Climate Change in Sulawesi Area 35

5.1.5 Climate Change in Nusatenggara Area 35

5.1.6 Climate Change in Moluccas Area 36

5.1.7 Climate Change in Papua Area 36

5.2 Additional Notes 36

5.2.1 Long Term Climate Variability 36

5.2.2 Climate Projection Uncertainties 38